* Sometimes variable length *inside* the server
* has maximum length on the wire.
*/
- if (value->enumv && value->enumv->flags.length) {
- if (max < value->enumv->flags.length) {
- FR_DBUFF_IN_MEMCPY_RETURN(&work_dbuff, (uint8_t const *)value->datum.ptr, max);
- FR_DBUFF_MEMSET_RETURN(&work_dbuff, 0, value->enumv->flags.length - max);
- return fr_dbuff_set(dbuff, &work_dbuff);
- }
+ if (value->enumv) {
+ if (value->enumv->flags.length) {
+ /*
+ * The field is fixed size, and the data is smaller than that, We zero-pad the field.
+ */
+ if (max < value->enumv->flags.length) {
+ FR_DBUFF_IN_MEMCPY_RETURN(&work_dbuff, (uint8_t const *)value->datum.ptr, max);
+ FR_DBUFF_MEMSET_RETURN(&work_dbuff, 0, value->enumv->flags.length - max);
+ return fr_dbuff_set(dbuff, &work_dbuff);
+
+ } else if (max > value->enumv->flags.length) {
+ /*
+ * Truncate the input to the maximum allowed length.
+ */
+ max = value->enumv->flags.length;
+ }
- /*
- * Truncate the input to the maximum allowed length.
- */
- max = value->enumv->flags.length;
+ } else if (da_is_length_field(value->enumv)) {
+ /*
+ * Truncate the output to the max allowed for this field and encode the length.
+ */
+ if (value->enumv->flags.subtype == FLAG_LENGTH_UINT8) {
+ if (max > 255) max = 255;
+ fr_dbuff_in(&work_dbuff, (uint8_t) max);
+
+ } else if (value->enumv->flags.subtype == FLAG_LENGTH_UINT16) {
+ if (max > 65536) max = 65535;
+ fr_dbuff_in(&work_dbuff, (uint16_t) max);
+
+ } else {
+ return -1;
+ }
+ }
}
FR_DBUFF_IN_MEMCPY_RETURN(&work_dbuff, (uint8_t const *)value->datum.ptr, max);
min, len);
return -(len);
}
+
+ /*
+ * For array entries, we only decode one value at a time.
+ */
if (len > max) {
- fr_strerror_printf("Found trailing garbage parsing type \"%s\". "
- "Expected length <= %zu bytes, got %zu bytes",
- fr_type_to_str(type),
+ if (!enumv->flags.array) {
+ fr_strerror_printf("Found trailing garbage parsing type \"%s\". "
+ "Expected length <= %zu bytes, got %zu bytes",
+ fr_type_to_str(type),
max, len);
- return -(max);
+ return -(max);
+ }
+
+ len = max;
}
- switch (type) {
- case FR_TYPE_STRING:
- if (fr_value_box_bstrndup_dbuff(ctx, dst, enumv, &work_dbuff, len, tainted) < 0) {
- return FR_VALUE_BOX_NET_OOM;
+ /*
+ * String / octets are special.
+ */
+ if (fr_type_is_variable_size(type)) {
+ size_t newlen = len;
+ size_t offset = 0;
+
+ /*
+ * Decode fixed-width fields.
+ */
+ if (enumv->flags.length) {
+ newlen = enumv->flags.length;
+
+ } else if (da_is_length_field(enumv)) {
+ /*
+ * Or fields with a length prefix.
+ */
+ if (enumv->flags.subtype == FLAG_LENGTH_UINT8) {
+ uint8_t num;
+
+ FR_DBUFF_OUT_RETURN(&num, &work_dbuff);
+ newlen = num;
+ offset = 1;
+
+ } else if (enumv->flags.subtype == FLAG_LENGTH_UINT16) {
+ uint16_t num;
+
+ FR_DBUFF_OUT_RETURN(&num, &work_dbuff);
+ newlen = num;
+ offset = 2;
+
+ } else {
+ return -1;
+ }
}
- return fr_dbuff_set(dbuff, &work_dbuff);
- case FR_TYPE_OCTETS:
- if (fr_value_box_memdup_dbuff(ctx, dst, enumv, &work_dbuff, len, tainted) < 0) {
- return FR_VALUE_BOX_NET_OOM;
+ /*
+ * If we need more data than exists, that's an error.
+ *
+ * Otherwise, bound the decoding to the count we found.
+ */
+ if (newlen > len) return -(newlen + offset);
+ len = newlen;
+
+ switch (type) {
+ case FR_TYPE_STRING:
+ if (fr_value_box_bstrndup_dbuff(ctx, dst, enumv, &work_dbuff, len, tainted) < 0) {
+ return FR_VALUE_BOX_NET_OOM;
+ }
+ return fr_dbuff_set(dbuff, &work_dbuff);
+
+ case FR_TYPE_OCTETS:
+ if (fr_value_box_memdup_dbuff(ctx, dst, enumv, &work_dbuff, len, tainted) < 0) {
+ return FR_VALUE_BOX_NET_OOM;
+ }
+ return fr_dbuff_set(dbuff, &work_dbuff);
+
+ default:
+ return -1;
}
- return fr_dbuff_set(dbuff, &work_dbuff);
- default:
- break;
}
/*
uint8_t const *data, size_t data_len, void *decode_ctx);
static ssize_t decode_value(TALLOC_CTX *ctx, fr_pair_list_t *out, fr_dict_attr_t const *parent,
- uint8_t const *data, size_t data_len, void *decode_ctx, bool exact);
+ uint8_t const *data, size_t data_len, void *decode_ctx);
static ssize_t decode_dns_labels(TALLOC_CTX *ctx, fr_pair_list_t *out,
fr_dict_attr_t const *parent,
if (parent->flags.array) return decode_array(ctx, out, parent, data, data_len, decode_ctx);
- return decode_value(ctx, out, parent, data, data_len, decode_ctx, true);
+ return decode_value(ctx, out, parent, data, data_len, decode_ctx);
}
/*
* Decode ONE value into a VP
*/
static ssize_t decode_value(TALLOC_CTX *ctx, fr_pair_list_t *out, fr_dict_attr_t const *da,
- uint8_t const *data, size_t data_len, void *decode_ctx, bool exact)
+ uint8_t const *data, size_t data_len, void *decode_ctx)
{
+ ssize_t slen;
fr_pair_t *vp;
uint8_t const *p = data;
uint8_t const *end = data + data_len;
+ bool exact = !da->flags.array;
FR_PROTO_TRACE("%s called to parse %zu bytes", __FUNCTION__, data_len);
FR_PROTO_HEX_DUMP(data, data_len, NULL);
* Structs create their own VP wrapper.
*/
if (da->type == FR_TYPE_STRUCT) {
- ssize_t slen;
-
slen = fr_struct_from_network(ctx, out, da, data, data_len, true,
decode_ctx, decode_value_trampoline, decode_tlv);
if (slen < 0) return slen;
if (da_is_split_prefix(da)) {
uint32_t ipaddr, mask;
- if (data_len != 8) goto raw;
+ if (data_len < 8) goto raw;
ipaddr = fr_nbo_to_uint32(p);
mask = fr_nbo_to_uint32(p + 4);
FALL_THROUGH;
default:
- {
- ssize_t ret;
-
- ret = fr_value_box_from_network(vp, &vp->data, vp->da->type, da,
+ slen = fr_value_box_from_network(vp, &vp->data, vp->da->type, da,
&FR_DBUFF_TMP(p, end - p), end - p, true);
- if (ret < 0) {
+ if (slen < 0) {
raw:
FR_PROTO_TRACE("decoding as unknown type");
if (fr_pair_to_unknown(vp) < 0) return -1;
break;
}
- if (exact && (ret != (end - p))) {
+ if (exact && (slen != (end - p))) {
talloc_free(vp);
goto raw;
}
- p += (size_t) ret;
- }
+ p += (size_t) slen;
+ break;
}
finish:
{
uint8_t const *p = data, *end = p + data_len;
ssize_t slen;
- size_t element_len;
FR_PROTO_HEX_DUMP(data, data_len, "decode_array");
"%s: Internal sanity check failed, attribute \"%s\" does not have array bit set",
__FUNCTION__, parent->name)) return PAIR_DECODE_FATAL_ERROR;
- /*
- * Fixed-size fields get decoded with a simple decoder.
- */
- element_len = fr_dhcpv4_attr_sizes[parent->type][0];
- if (!element_len) element_len = parent->flags.length;
-
- /*
- * Fix up insane nonsense.
- */
- if (da_is_split_prefix(parent)) element_len = 8;
-
- /*
- * Array of structs.
- */
- if (parent->type == FR_TYPE_STRUCT) {
- while (p < end) {
- slen = decode_value(ctx, out, parent, p, end - p, decode_ctx, false);
- if (slen < 0) return slen - (p - data);
- p += slen;
- }
-
- return data_len;
- }
-
- if (element_len > 0) {
- size_t num_elements = (end - p) / element_len;
-
- FR_PROTO_TRACE("decode_array %zu input expected %zd total (%zu elements * %zu size)",
- (size_t) (end - p), num_elements * element_len, num_elements, element_len);
-
- if ((num_elements * element_len) != (size_t) (end - p)) {
- raw:
- slen = decode_raw(ctx, out, parent, p, end - p , decode_ctx);
- if (slen < 0) return slen;
- return data_len;
- }
-
- while (p < end) {
- /*
- * Not enough room for one more element,
- * decode the last bit as raw data.
- */
- if ((size_t) (end - p) < element_len) goto raw;
-
- slen = decode_value(ctx, out, parent, p, element_len, decode_ctx, false);
- if (slen < 0) return slen - (p - data);
- if (!fr_cond_assert((size_t) slen == element_len)) return -(p - data);
-
- p += slen;
- }
-
- /*
- * We MUST have decoded the entire input. If
- * not, we ignore the extra bits.
- */
- return data_len;
- }
-
- /*
- * If the data is variable length i.e. strings or octets
- * there is a length field before each element.
- *
- * Note that we don't bother checking if the data type is
- * string or octets. There will only be issues if
- * someone edited the dictionaries and broke them.
- */
while (p < end) {
- if ((end - p) < 1) goto raw;
-
- element_len = *p;
-
- if ((size_t) (end - p) < (((size_t) element_len) + 1)) goto raw;
-
- p += 1;
- slen = decode_value(ctx, out, parent, p, element_len, decode_ctx, false);
+ slen = decode_value(ctx, out, parent, p, (end - p), decode_ctx);
if (slen < 0) return slen - (p - data);
+
p += slen;
}
+ /*
+ * We MUST have decoded the entire input. If
+ * not, we ignore the extra bits.
+ */
return data_len;
}
slen = decode_tlv(ctx, out, da, data + 2, len, decode_ctx);
} else {
- slen = decode_value(ctx, out, da, data + 2, len, decode_ctx, true);
+ slen = decode_value(ctx, out, da, data + 2, len, decode_ctx);
}
if (slen < 0) {
slen = decode_array(ctx, out, da, packet_ctx->buffer, q - packet_ctx->buffer, packet_ctx);
} else {
- slen = decode_value(ctx, out, da, packet_ctx->buffer, q - packet_ctx->buffer, packet_ctx, true);
+ slen = decode_value(ctx, out, da, packet_ctx->buffer, q - packet_ctx->buffer, packet_ctx);
}
if (slen <= 0) return slen;
fr_pair_t *vp;
fr_dict_attr_t const *da = da_stack->da[depth];
+ FR_PROTO_STACK_PRINT(da_stack, depth);
+
if (!fr_cond_assert_msg(da->flags.array,
"%s: Internal sanity check failed, attribute \"%s\" does not have array bit set",
__FUNCTION__, da->name)) return PAIR_ENCODE_FATAL_ERROR;
- /*
- * DNS labels have internalized length, so we don't need
- * length headers.
- */
- if ((da->type == FR_TYPE_STRING) && da_is_dns_label(da)) {
- while (fr_dbuff_extend(&work_dbuff)) {
- vp = fr_dcursor_current(cursor);
-
- /*
- * DNS labels get a special encoder. DNS labels are generally not compressed in
- * DHCPv4. But sometimes are compressed. Based on whatever the RFC author
- * decided was a good idea that day.
- */
- slen = fr_dns_label_from_value_box_dbuff(&work_dbuff, false, &vp->data, NULL);
- if (slen <= 0) return PAIR_ENCODE_FATAL_ERROR;
-
- vp = fr_dcursor_next(cursor);
- if (!vp || (vp->da != da)) break; /* Stop if we have an attribute of a different type */
- }
-
- return fr_dbuff_set(dbuff, &work_dbuff);
- }
-
while (fr_dbuff_extend(&work_dbuff)) {
- bool len_field = false;
fr_dbuff_t element_dbuff = FR_DBUFF(&work_dbuff);
- /*
- * If the data is variable length i.e. strings or octets
- * we need to include an 8-bit length field before each element.
- *
- * The struct encoder adds it's own length field.
- */
- if (da_is_length_field(da) && (da->type != FR_TYPE_STRUCT)) {
- len_field = true;
- FR_DBUFF_ADVANCE_RETURN(&element_dbuff, sizeof(uint8_t)); /* Make room for the length field */
- }
-
slen = encode_value(&element_dbuff, da_stack, depth, cursor, encode_ctx);
if (slen < 0) return slen;
- if (slen > UINT8_MAX) return PAIR_ENCODE_FATAL_ERROR;
- /*
- * Ensure we always create elements of the correct length.
- * This is mainly for fixed length octets type attributes
- * containing one or more keys.
- */
- if (da->flags.length && (size_t)slen < da->flags.length) {
- FR_DBUFF_MEMSET_RETURN(&element_dbuff, 0, da->flags.length - slen);
- }
-
- /*
- * Populate the length field
- */
- else if (len_field) fr_dbuff_in(&work_dbuff, (uint8_t) slen);
fr_dbuff_set(&work_dbuff, &element_dbuff);
vp = fr_dcursor_current(cursor);
* Address MAY be shorter than 16 bytes.
*/
case FR_TYPE_IPV6_PREFIX:
- if ((data_len == 0) || (data_len > (1 + sizeof(vp->vp_ipv6addr)))) {
+ if (data_len == 0) {
raw:
return decode_raw(ctx, out, parent, data, data_len, decode_ctx);
};
/*
- * Structs used fixed length fields
+ * Structs used fixed length IPv6 addressews.
*/
if (parent->parent->type == FR_TYPE_STRUCT) {
- if (data_len != (1 + sizeof(vp->vp_ipv6addr))) goto raw;
+ if (data_len != (1 + sizeof(vp->vp_ipv6addr))) {
+ goto raw;
+ }
vp = fr_pair_afrom_da(ctx, parent);
if (!vp) return PAIR_DECODE_OOM;
vp->vp_ip.af = AF_INET6;
vp->vp_ip.scope_id = 0;
vp->vp_ip.prefix = data[0];
- memcpy(&vp->vp_ipv6addr, data + 1, data_len - 1);
+ memcpy(&vp->vp_ipv6addr, data + 1, sizeof(vp->vp_ipv6addr));
+ slen = 1 + sizeof(vp->vp_ipv6addr);
break;
}
* If we have a /64 prefix but only 7 bytes of
* address, that's an error.
*/
- if (fr_bytes_from_bits(prefix_len) > (data_len - 1)) goto raw;
+ slen = fr_bytes_from_bits(prefix_len);
+ if ((size_t) slen > (data_len - 1)) {
+ goto raw;
+ }
vp = fr_pair_afrom_da(ctx, parent);
if (!vp) return PAIR_DECODE_OOM;
vp->vp_ip.af = AF_INET6;
vp->vp_ip.prefix = prefix_len;
- memcpy(&vp->vp_ipv6addr, data + 1, data_len - 1);
+ memcpy(&vp->vp_ipv6addr, data + 1, slen);
+
+ slen++;
break;
/*
vp = fr_pair_afrom_da(ctx, parent);
if (!vp) return PAIR_DECODE_OOM;
- if (fr_value_box_from_network(vp, &vp->data, vp->da->type, vp->da,
- &FR_DBUFF_TMP(data, data_len), data_len, true) < 0) {
+ slen = fr_value_box_from_network(vp, &vp->data, vp->da->type, vp->da,
+ &FR_DBUFF_TMP(data, data_len), data_len, true);
+ if (slen < 0) {
talloc_free(vp);
goto raw;
}
case FR_TYPE_STRUCT:
slen = fr_struct_from_network(ctx, out, parent, data, data_len, false,
decode_ctx, decode_value_trampoline, decode_tlv_trampoline);
- if (slen < 0) return slen;
+ if (slen < 0) goto raw;
+
+ if (parent->flags.array) return slen;
return data_len;
case FR_TYPE_GROUP:
* here.
*/
slen = decode_tlvs(vp, &vp->vp_group, fr_dict_root(dict_dhcpv6), data, data_len, decode_ctx);
+ if (slen < 0) {
+ talloc_free(vp);
+ return slen;
+ goto raw;
+ }
+ break;
+
+ case FR_TYPE_IPV6_ADDR:
+ vp = fr_pair_afrom_da(ctx, parent);
+ if (!vp) return PAIR_DECODE_OOM;
+
+ /*
+ * Limit the IPv6 address to 16 octets, with no scope.
+ */
+ if (data_len < sizeof(vp->vp_ipv6addr)) goto raw;
+
+ slen = fr_value_box_from_network(vp, &vp->data, vp->da->type, vp->da,
+ &FR_DBUFF_TMP(data, sizeof(vp->vp_ipv6addr)), sizeof(vp->vp_ipv6addr), true);
if (slen < 0) {
talloc_free(vp);
goto raw;
vp = fr_pair_afrom_da(ctx, parent);
if (!vp) return PAIR_DECODE_OOM;
- if (fr_value_box_from_network(vp, &vp->data, vp->da->type, vp->da,
- &FR_DBUFF_TMP(data, data_len), data_len, true) < 0) {
+ slen = fr_value_box_from_network(vp, &vp->data, vp->da->type, vp->da,
+ &FR_DBUFF_TMP(data, data_len), data_len, true);
+ if (slen < 0) {
talloc_free(vp);
goto raw;
}
break;
}
+ /*
+ * The input is larger than the decoded value, re-do it as a raw attribute.
+ */
+ if (!parent->flags.array && ((size_t) slen < data_len)) {
+ talloc_free(vp);
+ goto raw;
+ }
+
vp->vp_tainted = true;
fr_pair_append(out, vp);
+
+ if (parent->flags.array) return slen;
+
return data_len;
}
{
uint8_t const *p = data, *end = p + data_len;
ssize_t slen;
- size_t element_len;
FR_PROTO_HEX_DUMP(data, data_len, "decode_array");
"%s: Internal sanity check failed, attribute \"%s\" does not have array bit set",
__FUNCTION__, parent->name)) return PAIR_DECODE_FATAL_ERROR;
- /*
- * Fixed-size fields get decoded with a simple decoder.
- */
- element_len = fr_dhcpv6_attr_sizes[parent->type][0];
- if (!element_len) element_len = parent->flags.length;
-
- if (element_len > 0) {
- size_t num_elements = (end - p) / element_len;
-
- FR_PROTO_TRACE("decode_array %zu input expected %zd total (%zu elements * %zu size)",
- (size_t) (end - p), num_elements * element_len, num_elements, element_len);
-
- if ((num_elements * element_len) != (size_t) (end - p)) {
- raw:
- slen = decode_raw(ctx, out, parent, p, end - p , decode_ctx);
- if (slen < 0) return slen;
- return data_len;
- }
-
- while (p < end) {
- /*
- * Not enough room for one more element,
- * decode the last bit as raw data.
- */
- if ((size_t) (end - p) < element_len) goto raw;
-
- slen = decode_value(ctx, out, parent, p, element_len, decode_ctx);
- if (slen < 0) return slen;
- if (!fr_cond_assert((size_t) slen == element_len)) return -(p - data);
-
- p += slen;
- }
-
- /*
- * We MUST have decoded the entire input. If
- * not, we ignore the extra bits.
- */
- return data_len;
- }
-
- /*
- * If the data is variable length i.e. strings or octets
- * there is a length field before each element.
- *
- * Note that we don't bother checking if the data type is
- * string or octets. There will only be issues if
- * someone edited the dictionaries and broke them.
- *
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-...-+-+-+-+-+-+-+
- * | text-len | String |
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-...-+-+-+-+-+-+-+
- */
while (p < end) {
- if ((end - p) < 2) goto raw;
-
- element_len = fr_nbo_to_uint16(p);
-
- if ((size_t) (end - p) < (((size_t) element_len) + 2)) goto raw;
-
- p += 2;
- slen = decode_value(ctx, out, parent, p, element_len, decode_ctx);
+ slen = decode_value(ctx, out, parent, p, (end - p), decode_ctx);
if (slen < 0) return slen;
+
p += slen;
}
+ /*
+ * We MUST have decoded the entire input. If
+ * not, we ignore the extra bits.
+ */
return data_len;
}
option = DHCPV6_GET_OPTION_NUM(data);
len = DHCPV6_GET_OPTION_LEN(data);
if (len > (data_len - 4)) {
- fprintf(stderr, "FAIL %d\n", __LINE__);
fr_strerror_printf("%s: Option overflows input. "
"Optional length must be less than %zu bytes, got %zu bytes",
__FUNCTION__, data_len - 4, len);
"%s: Internal sanity check failed, attribute \"%s\" does not have array bit set",
__FUNCTION__, da->name)) return PAIR_ENCODE_FATAL_ERROR;
- /*
- * DNS labels have internalized length, so we don't need
- * length headers.
- */
- if ((da->type == FR_TYPE_STRING) && !da->flags.extra && da->flags.subtype){
- while (fr_dbuff_extend(&work_dbuff)) {
- vp = fr_dcursor_current(cursor);
-
- /*
- * DNS labels get a special encoder. DNS labels
- * MUST NOT be compressed in DHCP.
- *
- * https://tools.ietf.org/html/rfc8415#section-10
- */
- slen = fr_dns_label_from_value_box_dbuff(&work_dbuff, false, &vp->data, NULL);
- if (slen <= 0) return PAIR_ENCODE_FATAL_ERROR;
-
- vp = fr_dcursor_next(cursor);
- if (!vp || (vp->da != da)) break; /* Stop if we have an attribute of a different type */
- }
-
- return fr_dbuff_set(dbuff, &work_dbuff);
- }
-
while (fr_dbuff_extend(&work_dbuff)) {
- bool len_field = false;
fr_dbuff_t element_dbuff = FR_DBUFF(&work_dbuff);
- /*
- * If the data is variable length i.e. strings or octets
- * we need to include a length field before each element.
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-...-+-+-+-+-+-+-+
- * | text-len | String |
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-...-+-+-+-+-+-+-+
- *
- * The struct encoder already adds the length field.
- */
- if (da_is_length_field(da) && (da->type != FR_TYPE_STRUCT)) {
- len_field = true;
- FR_DBUFF_ADVANCE_RETURN(&element_dbuff, sizeof(uint16_t)); /* Make room for the length field */
- }
-
slen = encode_value(&element_dbuff, da_stack, depth, cursor, encode_ctx);
if (slen < 0) return slen;
- if (slen > UINT16_MAX) return PAIR_ENCODE_FATAL_ERROR;
- /*
- * Ensure we always create elements of the correct length.
- * This is mainly for fixed length octets type attributes
- * containing one or more keys.
- */
- if (da->flags.length && (size_t)slen < da->flags.length) {
- FR_DBUFF_MEMSET_RETURN(&element_dbuff, 0, da->flags.length - slen);
- }
-
- /*
- * Populate the length field
- */
- else if (len_field) fr_dbuff_in(&work_dbuff, (uint16_t) slen);
fr_dbuff_set(&work_dbuff, &element_dbuff);
vp = fr_dcursor_current(cursor);
# But if we don't have enough data, the whole thing is corrupt.
#
decode-pair 03 07 7f 00 00 01 7f 00 00
-match raw.Router-Address = 0x7f0000017f0000
+match Router-Address = 127.0.0.1, raw.Router-Address = 0x7f0000
count
match 21
decode-pair -
match User-Class = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxy"
-# too long results in "not encoded".
-# @todo - do we want to just truncate?
+# too long results just get truncated
encode-pair User-Class = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx123"
-match
+match 4d ff ff 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 4d 01 78
# Many longer ones
encode-pair User-Class = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx", User-Class = "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa", User-Class = "!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!"
projects / thirdparty / freeradius-server.git / commitdiff
